package Syntax::Highlight::HTML; use strict; use HTML::Parser; { no strict; $VERSION = '0.04'; @ISA = qw(HTML::Parser); } =head1 NAME Syntax::Highlight::HTML - Highlight HTML syntax =head1 VERSION Version 0.04 =cut my %classes = ( declaration => 'h-decl', # declaration process => 'h-pi', # process instruction comment => 'h-com', # comment angle_bracket => 'h-ab', # the characters '<' and '>' as tag delimiters tag_name => 'h-tag', # the tag name of an element attr_name => 'h-attr', # the attribute name attr_value => 'h-attv', # the attribute value entity => 'h-ent', # any entities: é « line_number => 'h-lno', # line number ); my %defaults = ( pre => 1, # add 
...
around the result? (default: yes) nnn => 0, # add line numbers (default: no) ); =head1 SYNOPSIS use Syntax::Highlight::HTML; my $highlighter = new Syntax::Highlight::HTML; $output = $highlighter->parse($html); If C<$html> contains the following HTML fragment:
some word
the description of the word. Plus some reference towards another definition.
then the resulting HTML contained in C<$output> will render like this: =begin html 
    <!-- a description list -->
    <dl compact="compact">
      <dt>some word</dt>
      <dd>the description of the word. Plus some <a href="/definitions/other_word"
      >reference</a> towards another definition. </dd>
    </dl>
=end html =head1 DESCRIPTION This module is designed to take raw HTML input and highlight it (using a CSS stylesheet, see L<"Notes"> for the classes). The returned HTML code is ready for inclusion in a web page. It is intented to be used as an highlighting filter, and as such does not reformat or reindent the original HTML code. =head1 METHODS =over 4 =item new() The constructor. Returns a C object, which derives from C. As such, any C method can be called on this object (that is, expect for C which is overloaded here). B =over 4 =item * C - Activate line numbering. Default value: 0 (disabled). =item * C
 - Surround result by C<< 
...
 >> tags. Default value: 1 (enabled). 

=back

B

To avoid surrounding the result by the C<< 
...
 >> tags:

    my $highlighter = Syntax::Highlight::HTML->new(pre => 0);

=cut

sub new {
    my $self = __PACKAGE__->SUPER::new(
        # API version
        api_version      => 3, 

        # Options
        case_sensitive   => 1, 
        attr_encoded     => 1, 

        # Handlers
        declaration_h    => [ \&_highlight_tag,  'self, event, tagname, attr, text' ], 
        process_h        => [ \&_highlight_tag,  'self, event, tagname, attr, text' ], 
        comment_h        => [ \&_highlight_tag,  'self, event, tagname, attr, text' ], 
        start_h          => [ \&_highlight_tag,  'self, event, tagname, attr, text' ], 
        end_h            => [ \&_highlight_tag,  'self, event, tagname, attr, text' ], 
        text_h           => [ \&_highlight_text, 'self, text' ], 
        default_h        => [ \&_highlight_text, 'self, text' ], 
    );
    
    my $class = ref $_[0] || $_[0]; shift;
    bless $self, $class;
    
    $self->{options} = { %defaults };
    
    my %args = @_;
    for my $arg (keys %defaults) {
        $self->{options}{$arg} = $args{$arg} if defined $args{$arg}
    }
    
    $self->{output} = '';
    
    return $self
}

=item parse()

Parse the HTML code given in argument and returns the highlighted HTML code, 
ready for inclusion in a web page. 

B

    $highlighter->parse("
Hello, world.
");

=cut

sub parse {
    my $self = shift;
    
    ## parse the HTML fragment
    $self->{output} = '';
    $self->SUPER::parse($_[0]);
    $self->eof;
    
    ## add line numbering?
    if($self->{options}{nnn}) {
        my $i = 1;
        $self->{output} =~ s|^|@{[sprintf '%3d', $i++]} |gm;
    }
    
    ## add 
...
?
    $self->{output} = "
\n" . $self->{output} . "
\n" if $self->{options}{pre};
    
    return $self->{output}
}

=back

=head2 Internals Methods

The following methods are for internal use only. 

=over 4

=item _highlight_tag()

C tags handler: highlights a tag. 

=cut

sub _highlight_tag {
    my $self = shift;
    my $event = shift;
    my $tagname = shift;
    my $attr = shift;
    
    $_[0] =~ s|&([^;]+;)|&$1|g;
    
    if($event eq 'declaration' or $event eq 'process' or $event eq 'comment') {
        $_[0] =~ s//>/g;
        $self->{output} .= qq|| . $_[0] . ''
    
    } else {
        $_[0] =~ s|^<$tagname|<$tagname|;
        $_[0] =~ s|^$tagname|;
        $_[0] =~ s|^<(/?)|<$1|;
        $_[0] =~ s|(/?)>$|$1>|;
        
        for my $attr_name (keys %$attr) {
            next if $attr_name eq '/';
            $_[0] =~ s{$attr_name=(["'])\Q$$attr{$attr_name}\E\1}
            {$attr_name=$1$$attr{$attr_name}$1}
        }
        
        $self->{output} .= $_[0];
    }
}

=item _highlight_text()

C text handler: highlights text. 

=cut

sub _highlight_text {
    my $self = shift;
    $_[0] =~ s|&([^;]+;)|&$1|g;
    $self->{output} .= $_[0];
}

=back

=head1 NOTES

The resulting HTML uses CSS to colourize the syntax. Here are the classes 
that you can define in your stylesheet. 

=over 4

=item *

C<.h-decl> - for a markup declaration; in a HTML document, the only 
markup declaration is the C, like: 
C<<  >>

=item *

C<.h-pi> - for a process instruction like C<<  >>
or C<<  >>

=item *

C<.h-com> - for a comment, C<<  >>

=item *

C<.h-ab> - for the characters C<< '<' >> and C<< '>' >> as tag delimiters

=item *

C<.h-tag> - for the tag name of an element

=item *

C<.h-attr> - for the attribute name

=item *

C<.h-attv> - for the attribute value

=item *

C<.h-ent> - for any entities: C<é> C<«>

=item *

C<.h-lno> - for the line numbers

=back

An example stylesheet can be found in F.

=head1 EXAMPLE

Here is an example of generated HTML output. It was generated with the 
script F. 

The following HTML fragment (which is the beginning of 
L)

    
    
     
      
      
      search.cpan.org: Sébastien Aperghis-Tramoni
     
     
    

    

     Home
    · Authors
    · Recent
    · News
    · Mirrors
    · FAQ
    · Feedback
    

    

    
    
in  
    


will be rendered like this (using the CSS stylesheet F): 

=begin html


  1 <!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN">
  2 <html>
  3  <head>
  4   <meta http-equiv="Content-Type" content="text/html; charset=iso-8859-1">
  5   <link rel="stylesheet" href="/s/style.css" type="text/css">
  6   <title>search.cpan.org: S&#233;bastien Aperghis-Tramoni</title>
  7  </head>
  8  <body id="cpansearch">
  9 <center><div class="logo"><a href="/"><img src="/s/img/cpan_banner.png" alt="CPAN"></a></div></center>
 10 <div class="menubar">
 11  <a href="/">Home</a>
 12 &middot; <a href="/author/">Authors</a>
 13 &middot; <a href="/recent">Recent</a>
 14 &middot; <a href="/news">News</a>
 15 &middot; <a href="/mirror">Mirrors</a>
 16 &middot; <a href="/faq.html">FAQ</a>
 17 &middot; <a href="/feedback">Feedback</a>
 18 </div>
 19 <form method="get" action="/search" name="f" class="searchbox">
 20 <input type="text" name="query" value="" size="35">
 21 <br>in <select name="mode">
 22  <option value="all">All</option>
 23  <option value="module" >Modules</option>
 24  <option value="dist" >Distributions</option>
 25  <option value="author" >Authors</option>
 26 </select>&nbsp;<input type="submit" value="CPAN Search">
 27 </form>



=end html

=head1 CAVEATS

C relies on C for parsing the HTML 
and therefore suffers from the same limitations. 

=head1 SEE ALSO

L

=head1 AUTHORS

SEbastien Aperghis-Tramoni, Esebastien@aperghis.netE

=head1 BUGS

Please report any bugs or feature requests to
C, or through the web interface at
L. 
I will be notified, and then you'll automatically be notified of 
progress on your bug as I make changes.

=head1 COPYRIGHT & LICENSE

Copyright (C)2004 SEbastien Aperghis-Tramoni, All Rights Reserved.

This program is free software; you can redistribute it and/or modify it
under the same terms as Perl itself.

=cut

1; # End of Syntax::Highlight::HTML